Polarization Camera Surface Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer vision techniques fail to reliably detect surface defects on optically challenging surfaces, such as scratches on glass or dents in glossy paint, due to low contrast between defect and background colors, making it difficult to improve manufacturing efficiency and reduce costs.
Innovation Solution
The use of polarization-enhanced imaging with polarization cameras capturing frames at different polarizations, extracting tensors in polarization representation spaces, and employing machine learning models like convolutional neural networks to detect surface characteristics, including defects, by analyzing the degree of linear polarization and angle of linear polarization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional computer vision techniques are used to detect surface defects, then the system is simple and easy to operate, but the detection reliability is poor due to low contrast between defect and background colors
Solution Approach 1:
The patent transitions from conventional 2D intensity-based imaging to 4D polarization imaging by adding two dimensional parameters (degree of linear polarization and angle of linear polarization) to the traditional intensity and spatial dimensions. This dimensional expansion enables detection of surface defects through polarization contrast rather than relying solely on color contrast, thereby improving detection reliability on optically challenging surfaces.
Solution Approach 2:
The patent changes the physical parameters used for defect detection from intensity/color parameters to polarization parameters. By measuring the degree of linear polarization (DOLP) and angle of linear polarization (AOLP) in addition to intensity, the system creates new contrast mechanisms that are sensitive to surface geometry and material properties, enabling reliable defect detection where conventional intensity-based methods fail.
2Measurement precision
If polarization-enhanced imaging with multiple polarization channels is used, then the surface defect detection accuracy is improved, but the data processing complexity and computational requirements increase
Solution Approach 1:
The patent segments the polarization data processing into distinct functional modules: (1) extraction of degree of linear polarization (DOLP) maps, (2) extraction of angle of linear polarization (AOLP) maps, and (3) fusion with intensity data for defect detection. This modular segmentation of the complex polarization processing pipeline reduces overall computational complexity by organizing operations into manageable, independent stages that can be processed efficiently.
Solution Approach 2:
The patent introduces intermediate polarization parameter maps (DOLP and AOLP) as mediators between the raw polarization images and the final defect detection results. These intermediate representations condense the complex polarization information into two key physical parameters that are more computationally efficient to process than full polarization tensors, while still preserving the essential contrast information needed for accurate defect detection.
3Productivity
If conventional imaging is used, then the equipment cost is low, but the manufacturing efficiency and quality assurance capability are limited
Solution Approach 1:
The patent implements a polarization camera system that captures multiple types of information (intensity, DOLP, AOLP) simultaneously from a single optical setup. This multi-functional capability allows the same hardware to perform both conventional intensity-based inspection and polarization-based defect detection, eliminating the need for separate imaging systems and maximizing the utility of the equipment investment for quality assurance.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach significantly improves the detection of surface defects on optically challenging surfaces, enhancing manufacturing efficiency and reducing costs by providing accurate and robust characterization of surface features that were previously difficult to detect.
Implementation Method 1
receiving one or more polarization raw frames of a surface of a physical object, the polarization raw frames being captured at different polarizations by a polarization camera including a polarizing filter
Data Source
AI summary
A computer-implemented method for surface modeling includes: receiving one or more polarization raw frames of a surface of a physical object, the polarization raw frames being captured with a polarizing filter at different linear polarization angles; extracting one or more first tensors in one or more polarization representation spaces from the polarization raw frames; and detecting a surface characteristic of the surface of the physical object based on the one or more first tensors in the one or more polarization representation spaces.